D2 DUsing Veritas Netbackup

Online Disk-to-Disk Backup with
Veritas NetBackup and VSS
This solution guide describes how to configure Intransa Large Scale IP SAN storage System
with Veritas NetBackup and integrate with VSS for best performance as destination for
backups. It also compares the benefits and performance characteristics of backup-to-disk
verses backup-to-tape implementations.
© 2004-2005 Intransa, Incorporated. All rights reserved worldwide.
Unauthorized duplication or reproduction of this document by any means, including electronic, reprographic, or
any other form without written permission from Intransa, Inc., is strictly forbidden.
Use of this manual and the software program is subject to the terms of the License Agreement between you and
Intransa, Inc. You should not use the software until you have read the License Agreement. By using the
software program, you signify that you have read the License Agreement and accept its terms. For a complete
explanation of the terms of this agreement, see the License Agreement booklet.
Trademarks
The Intransa logo, Intransa Storage System, StorControl, StorControl Management Tool, IntraStor,
and the Simply Smarter logo are all registered trademarks of Intransa, Inc.
Microsoft® is a registered trademark of Microsoft Corporation. Windows™ is a trademark of Microsoft

Corporation.
Java® is a registered trademark Sun Microsystems, Inc.
Linux® is a registered trademark of Linus Torvalds.
Oracle® is a registered trademark of Oracle Corporation.
All other brand or product names are trademarks or registered trademarks of their respective companies or
organization. All names of people and companies used for examples in this manual are fictional. Any similarity
to names of real companies and real people is coincidental.
Revisions
Intransa reserves the right to revise this document and the information within at any time and as necessary
without the obligation to notify any person of revision.
Contact:
Intransa, Inc.
2870 Zanker Road Suite 200
San Jose, CA 95134-2114
Phone: 408.678.8600
Fax: 408.678.8800
Web: http://www.intransa.com
Online Disk-to-Disk Backup with Veritas NetBackup and VSS Intransa Inc.
Contents
1. Executive Summary .................................................................................................... 1
Overview.....................................................................................................................................1
Intransa Disk-to-Disk (D2D) Backup Strategy ......................................................................2
Conclusion ..................................................................................................................................4
2. Preface ........................................................................................................................... 5
Introduction................................................................................................................................5
Disk Backup Overview .......................................................................................................................5
Advantages of Backup-to-Disk ..........................................................................................................6
About Intransa ...........................................................................................................................8
Intransa – A brief profile.....................................................................................................................8
Intransa Approach to Solutions .........................................................................................................8
About Veritas ...........................................................................................................................10
Users ....................................................................................................................................................10
Other Useful Resources.....................................................................................................................11
3. Veritas NetBackup .................................................................................................... 12

Overview...................................................................................................................................12
About Advanced Client ....................................................................................................................14
Snapshots ............................................................................................................................................14
4. NetBackup Qualification Setup ............................................................................. 16

Getting Started .........................................................................................................................16
Assumptions.......................................................................................................................................17
Configurations used in the Qualification Environment ...............................................................19
5. Configuring IP SAN Volumes................................................................................ 22

Intransa Large Scale IP SAN Overview:...............................................................................22
What is an IP SAN? .................................................................................................................23
Solution Specific Configurations .....................................................................................................24
Windows File System Tuning ................................................................................................31
6. Installing Intransa VSS Provider ........................................................................... 34
7. Installing NetBackup Server and Client .............................................................. 39

NetBackup Client Installation................................................................................................58
Contents i
8. Testing Backup and Restore ................................................................................... 66

Defining Storage Unit .............................................................................................................66
Defining Backup Policy.....................................................................................................................68
Verifying the Snapshot operations ..................................................................................................80
9. Performance Benchmarking.................................................................................... 86
10. Summary................................................................................................................... 88
Conclusions ..............................................................................................................................88
Summary...................................................................................................................................88
11. Appendix .................................................................................................................. 90

A.VSS Concepts .......................................................................................................................90
A.1 What is VSS:.................................................................................................................................90
A.2 VSS Components ........................................................................................................................91
A.3 How VSS works ..........................................................................................................................91
B. Overview to VSS and VDS Design ...................................................................................93
C. Volume Shadow Copy Service..........................................................................................95
D. Tape Backups: Limitations ................................................................................................99
E. Essentials of Veritas NetBackup Administration .........................................................102
E.1 NetBackup Management..........................................................................................................103
E.2 NetBackup Configuration Wizards ........................................................................................109
E.3 Disk Staging Storage Unit Considerations.............................................................................110
E.4 Types of Backup ........................................................................................................................112
E.5 Configuring an Advanced Client Policy ................................................................................114
Reference and Acknowledgements ......................................................................... 116

References ...............................................................................................................................116
Acknowledgements...............................................................................................................116
Contents ii
1. Executive Summary
In a 24x7x365 world with geographic distribution of users, even shrinking back
up windows is becoming unacceptable, especially for mission-critical
applications. Enterprise environments need up-to-the-minute data protection
without sacrificing high availability. Especially in customer centric and service
centric organizations, every minute of backup restore is important. In case of
outage or disaster every minute of delay in putting back the operations translates
into opportunity loss in terms of money, customers and image. In some cases it
can be as serious as closing business.
Traditionally backup was taken in tape cartridges. However off late, disk
capacities have overtaken single tape cartridge capacities. There seems to be a lot
of interest by organizations to go for near line storage by deploying disk to disk
backup solutions rather than taking directly into tapes. This makes business and
economic sense in the whole run.
Overview
Backup to disk solutions provide tremendous benefits compare to traditional
backup-to-tape. These benefits are as follows:
Faster recovery of mission critical data
Faster restore from disk
Greater reliability of the backup media
Multiple backup and restore data streams to disk
Backup-to-disk is emerging as powerful solution for faster recovery of mission-

critical data and it is complementary to backup-to-tape for long term storage.
Most of the applications today require frequent retrievals of recently captured

data, as well as the need for backup of business critical data which is required to
be preserved and retrieved quickly and efficiently. Today most of the leading-
edge customers prefer using disk as the destination for nearline, faster backup as
backup windows are shrinking and importance of data is increasing every day.
1. Executive Summary - 1
Advanced Technology Attached (ATA) disk technology provides greater benefits

over traditional tape with the increasing performance and reducing cost of disk.
Disk-based backups will not replace tape, but will definitely shift tape into an
archival role. Both of these media will co-exist.
Today, Intransa recommends adopting a backup-to-disk solution when

performance and data reliability are the primary concerns of the customer.
Intransa Disk-to-Disk (D2D) Backup Strategy

Intransa has developed Large Scale IP SAN that is built around a set of modular
components to deliver exceptional value, low total cost of ownership, high
scalability, excellent performance, and built-in redundancy & reliability. These
requirements define IntraStor™ architecture, our foundation for delivering
block-level, distributed storage over IP-networks.
Intransa IntraStor™ is a 3-tiered distributed architecture that fully leverages the

ubiquity of IP infrastructure and the iSCSI protocol to deliver intelligent,
networked block-level storage. This offers the flexible scalability options in
scaling up the storage capacity or storage throughput I/O performance or both.
This solution of Disk-to-Disk backup is using Veritas NetBackup software and

Intransa VSS Provider software. Intransa offers the flexibility in performing the
backup from the snapshot. Intransa VSS Provider smartly takes care of creating
snapshot of the production data volume during the backup operations and
deletes the snapshot after completing the backup. This reduces the burden of
backup from the Production data volume and keeping it free only for the
production purpose. Intransa’s scalable architecture offers the unique
combination linear scalability in performance and capacity in pay as you grow
mode.
This paper gives an overview of how to deploy Disk-to-Disk Backup Solution on

Intransa Large Scale IP SAN using Veritas NetBackup and Intransa VSS
Provider.
The paper is lucidly divided into the following logical sections:
Section 2 is Preface gives introduction and brief need of the Solution. It also
covers brief profiles about Intransa and Veritas.
Section 3 provides the overview of Veritas NetBackup Software, its architecture,

basic functionality.
Section 4 provides the overview of Veritas NetBackup Qualification

configurations setup and brief specifications of the servers used in the
qualification environment.
Section 5 provides the overview of Intransa Large Scale IP SAN and step-by-step
guide in configuring the IP SAN Volumes for the Disk-to-Disk Solution.
Section 6 provides step-by-step instructions for installing and configuring the

Intransa VSS Provider software for the Disk-to-Disk backup Solution.
Section 7 provides step-by-step instructions for Installation and configuring

NetBackup Server and Clients for the Solution using the snapshot feature.
Section 8 provides brief about defining the Disk Storage Unit , Backup Policy for
using the snapshot feature for the solution and also provides the steps for
conducting the backup and restore operations
Sections 9 provides the performance benchmarks for the backup & restore tests
conducted.
Section 10 provides summary of the solution and the conclusions of the tests.
Section 11 is the appendix containing the additional information about the

components used in the solution.
Brief concepts of Microsoft Virtual Shadow Copy Service.
Brief essentials of Veritas NetBackup Software, backup types and some
useful guidelines and suggestions in configuring the solution.
Section 12 is about the references used for this solution document and
acknowledging the teams involved in formulating this solution.
Conclusion
The performance characteristics are very encouraging. It shows that backup
windows can typically shrink by 3 times or reduce back up times by 66 %
compared to the fastest LTO based solutions in the market.
Veritas NetBackup Software seamlessly gets integrated with the Intransa VSS
Provider. This gives immense benefit for configuring the Disk-to-Disk Backup
Solutions with Intransa IP SAN in using the automated snapshot feature.
A D2D solution of the nature proposed will benefit the IT staff to increase the
numbers of backups or simply ease the management and cost of backups. This
also insures that the backups are always near online and easily accessible.
2. Preface
Introduction
This solution guide explains implementing backup-to-disk technology with
Veritas NetBackup and the advantages over traditional tape based backup. These
implementations have been tested by Intransa Solution Engineering teams.
This document contains information relating to configuring Veritas NetBackup

and Intransa IP SAN, Intransa VSS provider to form the backup-to-disk solution.
It also compares performance with the traditional tape backups.
Intransa Solution Engineering teams performed tests using specific system

configurations. Due to variations in hardware, software, compressibility of data
system usage and other factors, backup rates achieved in some situation may be
different than those shown herein. All testing performed by the Solution
engineering teams used the most up-to-date backup software available at that
time. Other versions of these applications any yield other results.
Disk Backup Overview

Traditionally, backup software was created to write to a tape device. Today, most
backup software products also support write to disk, which means writing to
disk file in a file system. The file system may be on a Windows, Netware or
UNIX platform, depending on the backup server. Disk-based RAID
configurations enhance data protection beyond what tape can provide.
Backup and restore time using disk provides customers an advantage over using
traditional tape. Disk drives are random access devices and can instantly start to
transfer the files, whereas with tape, the tape must be loaded and then accessed,
increasing the overall time. File system overhead and tape with on-board
compression add more time lags.
Under normal conditions, I/O will always go to physical disk. Performances are
characterized by the physical disk rotational speed and seek time together with
the file system characteristics. Using a larger file system block size or allocation
unit size may improve performance of disk backups by using larger contiguous
2. Preface - 5
address space for files because the operating system allocates fewer groups of
contiguous sectors.
Benefits of IP SAN disk-based backups that are discussed in more detail in this
solution document include:
Random access characteristics of disk versus sequential access for tape

simultaneous capability to read/write to disk.
Multiple host data streams
Disk performance compared to tape performance with compression.
Aggregate performance
RAW disk performance
Advantages of Backup-to-Disk
Traditionally, tape has been the backup medium of choice, due to its cost-per-MB
advantages compared with disk. However the economics of disk narrowing that
gap. The advantages of using disk over with backup solutions can be grouped
into four major categories:
Backup performance
Restore performance
Media reliability and data availability
Overall IT efficiency.
The following sections summarize these benefits.
Backup Performance
IP SAN storage systems based backup are much faster than tape drives.
Some tape technologies respond to a minimal data stream by “shoe-
shining” or excessive positioning. Disks do not experience this behavior
as they are inherently random access.
2. Preface - 6
Restore Performance
Data Recovery from disks are much faster than tape. The difference can be
seconds or minutes, verses hours with tape.
Disk supports random and sequential access. Tape support sequential access
only. This enables faster access to data files, improving overall performance. If
data is on several tape cartridges, the following steps are required to restore the
data:
1. Each tape must be mounted by the library (up to a minute per tape)
2. Tape must load (30 seconds to a few minutes )
3. Tape must be positioned to the desired data ( an average access time is a
few minutes )
4. Tape must be rewound and unloaded ( 30 seconds to a few minutes )
5. Load the next tape and repeat cycle.
Time to first byte takes milliseconds for disk versus seconds to minutes to tape.
Media Reliability and Data Availability

Media-specific errors, including faulty tape media, are reduced. Disk
system RAID protection prevents data unavailability or data loss in the
event of a disk drive failure.
Tape handing is reduced or eliminated. Maintaining a set of tapes from a
tape library can be problematic and requires properly trained personnel.
Overall IT Efficiency
Disk does not require the tape handling /positioning and RAID protection
makes it inherently more reliable. There is less need to perform frequent
full backups. Fewer backups need to be performed, saving network and
CPU load.
Tape undergoes a technology shift every three years, so a conversion
process from old to new media must be undertaken at that interval. Disk
technology does not go through these types of transitions since the format
of the data is not changed as it is with tape technology.
New larger-capacity disk drive reduces floor space requirements
compared with equivalent-capacity tape libraries.
2. Preface - 7
About Intransa
Intransa – A brief profile
Founded in September 2000, Intransa is an innovator of enterprise-class,
intelligent IP storage products and software. Based on technology that leverages
the low cost and maturity of Ethernet technology, Intransa’s IP SAN solutions
dramatically reduce the cost and complexity of deploying scalable network
storage. Intransa’s products re-define network storage to deliver the first full-
featured IP SAN at a price attractive to organizations. Intransa believes that
network storage should be easy, dynamic and intuitive. By achieving these goals,
Intransa is enabling business customers to focus their resources onto mission-
critical projects to improve their business.
Intransa’s new IP SAN Storage System is a second-generation iSCSI-enabled,

networked storage solution that implements the IntraStor™ architecture. This
solution offers low cost, block-level, networked storage that can be deployed
anywhere on the IP network and managed centrally. The enterprise-class
features, such as fail-over and load balancing, support the high-availability needs
of online data. The policy-based virtual volume management and dynamic
volume expansion capabilities, in conjunction with the Intransa’s inexpensive
disk storage pool, simplify capacity management and improve capacity
utilization for online, nearline, and backup/restore staging applications.
Intransa’s snapshot capability allows many point-in-time virtual copies of data,
representing days and weeks of data, thereby obviating the need to
incrementally backup to tape on a daily basis.
Intransa Approach to Solutions

Intransa, with its partners, is focused and committed, to delivering Solutions to
market that allows customers to implement end-to-end storage solutions. The
Solutions are targeted to meet demanding customer needs.
2. Preface - 8
Intransa’s solution leverage on the simplicity, scalability and flexibility of the

IntraStor Architecture. IntraStor architecture is an enabler for a solutions-based
platform. The above figure gives a conceptual overview of the N-way Scaling in
performance and capacity.
The following approach is helping Intransa to rapidly build a growing and

satisfied client base all over the world.
Create a storage network as you would create your client/server (data)
network
Leverage your existing IP infrastructure for solutions such as Disaster
Recovery
Leverage existing IP expertise and resources in organization
Minimize the number of disparate networks that organizations have to
manage and support
Increase return on your investment via Simplicity of Scaling the IntraStor
architecture and solution
Intransa has various families of products like IP3000, IP5500 and IP7500 to
address storage, price, performance and availability requirements of the
customer. These scalable solutions use IP as a storage medium and can be scaled
to more than 50 TB. The N-way clustering of Storage Controllers help in scaling
2. Preface - 9
up the performance on demand. Intransa will continue to develop value add

features that allows our customers to take advantage of the IP network
investment that they have already made.
For more details about Intransa please visit http://www.intransa.com
About Veritas
VERITAS Software is a leading provider of software and services to enable utility
computing. In a utility computing model, IT resources are aligned with business
needs and business applications are delivered with optimal performance and
availability on top of a shared computing infrastructure, minimizing hardware
and labor costs.
The VERITAS NetBackup mission is to provide heterogeneous data protection

solutions ranging from the workgroup or departmental level to the enterprise
level.
The NetBackup Enterprise Server product provides powerful functionality,

flexibility, mainframe-caliber robustness, and high performance in large
enterprise environments.
The NetBackup Server product is a high-performance solution offering ease of

use and functionality that is appropriate to smaller work environments.
NetBackup contains many significant enhancements in the area of scalability,
application protection, clustering, advanced client and off-host backups, and
integration with VERITAS Backup Exec.
For more details about Veritas please visit http://www.veritas.com
Users
This guide is intended to be used by Backup, System and network
administrators. Throughout this document we assume basic understanding of:
Local Area Networks (LAN)
Ethernet and Ethernet-switching concepts
Storage Area Network (SAN)
Backup and System Administration
2. Preface - 10
Other Useful Resources

Intransa IP SAN Administrator’s Guide.
Intransa IP SAN Command Line Interface Guide.
Intransa IP SAN StorControl Management Tool Guide.
Veritas NetBackup Installation Guide
Veritas NetBackup System Administration Guide
Veritas NetBackup Advanced Client Quick Start Guide
2. Preface - 11
3. Veritas NetBackup
This section provides brief introduction of Veritas NetBackup software,
architecture, functionality and basics of snapshot.
Overview
Veritas NetBackup is based on client/server architecture. Each NetBackup client
and server belongs to a storage domain. A storage domain consists of a single
master Server, its associated “Media Servers” and “NetBackup clients”. The
“Master Server” controls and directs all NetBackup operations in its storage
domain. Each Media Server controls the backup devices it is connected to,
including direct-attach and SAN devices. A Media server can have only one
Master Server, but a Master Server can control more than one Media Server. The
NetBackup clients are any systems containing data to be backed up. A Master
Server can act as a Media Server, and are capable of being clients.
A NetBackup client is any system with data to be backed up. The client software
is tailored to the operating system on which it is installed. Normally, a client
operates under the control of the Master Server according to the rules and
schedules that an administer establishes. A backup client accesses the storage
media through a Media Server.
A NetBackup Media Server hosts one or more backup devices. Storage devices
on the same SAN can be shared between Media Servers using the Shared Storage
Option (SSO). The Master Server directs a client to send its data to a Media
Server for backup. SSO does not support disk devices.
A NetBackup Master Server is the manager of the storage domain. An

administrator can control all NetBackup functions in the storage domain from
the Master Server.
NetBackup provides high-performance backups and restores for a variety of

platforms, including Microsoft Windows, UNIX, and NetWare systems.
3. Veritas NetBackup - 12
Administrators can set up schedules for automatic, unattended backups for

clients anywhere in the network. These backups can be full or incremental and
are managed entirely by the NetBackup master server.
Users can start backups and restores from the computer where they are working.
A user can also archive files. An archive operation backs up a file and then
deletes it from the local disk if the backup is successful. Once started, user
operations are managed by the NetBackup server.
NetBackup’s Media Manager software manages the media and storage devices.
Robots require no intervention on the part of the administrator, operator, or the
user. Standalone drives (those not in a robot) that contain appropriate media also
require no intervention.
NetBackup includes both the server and client software:

Server software is on the computer that manages the storage devices.
Client software is on the computer whose data you want to back up. A server
also has client software and can be backed up like other clients.
NetBackup supports both master and media servers. The master server manages
the backups, archives, and restores. Media servers provide additional storage by
allowing NetBackup to use the storage devices that they control. Media servers
can also increase performance by distributing the network load.
During a backup or archive, the client sends backup data across the network to a
NetBackup server that has the type of storage specified for the client. The storage
requirement is specified during NetBackup configuration
During a restore, users can browse and then select the files and directories that
they want to recover. NetBackup finds the selected files and directories and
restores them to the disk on the client.
About Advanced Client

Advanced Client option is required for taking the snapshots and Veritas
NetBackup integration with Intransa VSS Provider.
Advanced Client combines the features of snapshot backup, FlashBackup, BLI

Agent, offhost backup, and Instant Recovery.
Snapshots
A snapshot is a disk image of the client’s data made almost instantaneously.
NetBackup backs up the data from the snapshot image, not directly from the
client’s primary data. This allows client operations and user access to continue
without interruption during the backup.
A snapshot image is required for all features of Advanced Client. A number of

methods are provided for creating snapshots. You can select the snapshot
method manually from the Policy dialog of the NetBackup Administration
Console, or allow NetBackup to select the method for you.
Snapshot Basics
Large active databases or file systems that must be available around-the-clock are
difficult to back up without incurring a penalty. Often, the penalty takes one of
two forms:
The entire database is taken offline or the file system is unmounted, to
allow time for the backup, resulting in suspension of service and
inconvenience to users.
The copy is made very quickly but produces an incomplete version of the
data, some transactions having failed to complete.
A solution to this problem is to create a snapshot of the data. This means

capturing the data at a particular instant, without causing significant client
downtime. The resulting capture or snapshot can be backed up without affecting
the performance or availability of the file system or database. Without a
complete, up-to-date snapshot of the data, a correct backup cannot be made.
When a backup is managed by a backup agent on a Storage network, the data to

backup must be contained in a snapshot. The backup agent can only access the
data by means of the raw physical disk. Once the data is captured as a snapshot,
the NetBackup client “maps” the logical representation of the data to its absolute
physical disk address.
More details about administering the NetBackup Server and Microsoft’s Volume
Shadow copy Service (VSS) frame work along with related technical details are
covered on Appendix.
4. NetBackup Qualification Setup

This section provides guidelines and in setting up the servers for Configuring
the IP SAN volumes and Configuring Veritas NetBackup Server and Client to
provide Disk-to-disk backup using Microsoft VSS framework. This section also
provides brief specifications of the servers used in the qualification environment.
Getting Started
Before beginning the setup process, make sure you have gathered the following
equipment and necessary information:
Intransa IP SAN along with Intransa VSS Provider Software.
The Veritas NetBackup Software Enterprise Edition (Version 5) with its
Advanced Client license.
Windows 2003 Server with ServicePack-1, a Gigabit NIC and Microsoft
iSCSI initiator. One other NIC for management interface.
Network settings specific to your network, such as: IP addresses,
Network setting, Netmask setting, and Gateway setting.
Network Switches with proper V-LAN configurations.
4. NetBackup Qualification Setup - 16

Assumptions
It is assumed that the following conditions exist before proceeding with the setup
procedure:
NetBackup Server (NS) and NetBackup Client (NC) are installed with
Windows 2003 Enterprise Edition with Updated Service Packs.
NS and NC are configured with Intel Gigabit Ethernet NIC with
Microsoft initiators to connect to the Host Access Network (HAN) of IP
SAN.
The setup machines have two network adapters:
One Gigabit adapter to connect to the iSCSI Network of IP SAN.
Second adapter to connect the NS, NC and other Client machine for
backup management traffic on LAN switch.
Gigabit Ethernet Switch is used to connect the NS and NC to Host Access
Network of IP SAN.
Jumbo frames are enabled on the Storage Data Network (SDN) GigE
switch.

The proposed D2D solution comprises of the following architecture:
The NetBackup Server (NS) is installed on an independent server and connected

IP SAN for taking the backup on the disk. The NetBackup Clients utilizes IP SAN
volumes to host the data.
The IP SAN provides the snapshot taking capability through built-in snapshot
feature in the StorControl Software and Intransa VSS Provider software installed
on the NC of which backup is to be taken.

NetBackup Server provides scheduling and management functionality for

initiating snapshots, scheduling the backup/restore operations, and deleting
snapshots. In the above architecture, the backup data traffic flow from the IP
SAN Snapshot Volume mounted on NC to the IP SAN netbackup volume.
The data backup occurs from snapshot volumes (provided by IP SAN) mounted
at the Client to a backup volume on the IP SAN.
Configurations used in the Qualification Environment

Following is the qualification environment configured for the evaluation.
Servers
The configurations used in qualification environment are two Pentium 4 servers
connected via Intel GigE Ethernet Adapter on LAN.
NetBackup Server (NS)

Pentium 4 server named as “Rasgulla” is configured as NetBackup Server.
NetBackup Client (NC)

Pentium 4 server named as “Jupiter” is configured as NetBackup Client.
Both machines are connected to Intransa large scale IP SAN storage via Gigabit
Ethernet NIC with Microsoft iSCSI initiator
Networks
There are three types of Networks involved in the configurations
1. LAN: Local Area Network

Gigabit Ethernet LAN switch:
It is used for establishing the connectivity between the Servers and LAN
Clients.
2. HAN: Host Access Network

Gigabit Ethernet HAN switch:
The connectivity between the IP SAN Host Access Network (HAN) and
NetBackup Servers is done using the HP Gigabit Switch (HP procurve
switch 2824) with no jumbo frame support.
3. SDN: Storage Data Network

Gigabit Ethernet SDN switch:
Storage Data Network (SDN) is comprises of ETH3 and ETH4 network
ports of storage controllers and connections from the each Disk

Enclosures. This is connected using Foundry EdgeIron 48G switch with

jumbo frame enabled on the switch.
Software
Windows 2003 Server Enterprise Edition with updated service pack and
NTFS file system is installed on NS & NC machines.
VERITAS NetBackup Server for Microsoft Windows Version 5.0
Intransa VSS Provider software.
Volumes
IP SAN Volume namely “netbackup” is configured using NTFS-Optimized
custom policy (which is described latter in the document) for taking the backup.
Also IP SAN Volume namely “source1” to “source 10” are created using simple
stripe policy and used as source volumes for backup.
Volume netbackup is assigned to Netbackup Server “Rasgulla” via Storage

Controller Module (SCM 1) and Source volumes are assigned to Netbackup
Client “Jupiter” via Storage Controller Module (SCM 2)
Data
Set of 10 GB streaming datasets are prepared on the NetBackup Client (Jupiter)
as source data for backup.

Qualification Configuration in brief is in given in Table 1 below.
Table 1 :
No. Item Configuration
NetBackup Server and Client

1 Processor Pentium 4, 2.4 GHz Server with single CPU
2 Memory 1 GB (DDR266)
3 Local Hard disk 40 GB ATA
4 Network card Intel Pro Ethernet NIC
5 Host Based Adapter Intel Pro GigE Ethernet NIC with Microsoft iSCSI
initiator
IP SAN : Intransa IP 5500
7 Storage Controller 2 X SC5500 in cluster mode with build 1.4.00
Module
8 Disk Enclosure 1 X DE5200 (16 X 250 GB Maxtor ATA disk drives)
9 Operating System Windows 2003 Server Enterprise Edition and

Service Pack 1
10 Host Access HP procurve switch 2824
Network
12 Storage Data Foundry EdgeIron 48G
Network
13 Veritas Veritas NetBackup Server 5.0
14 Intransa VSS Provider Software
15 Data 10 GB streaming data

5. Configuring IP SAN Volumes
Intransa Large Scale IP SAN Overview:
IP SAN overview
5. Configuring IP SAN Volumes - 22

What is an IP SAN?
The IP SAN is an iSCSI-based storage system. It provides a large, scalable, easily
managed pool of storage. The IP SAN is the best of both worlds: the block-level
abilities of a Storage Area Network (SAN), and the simplicity and cost benefits of
a Network Attached Storage (NAS) device. In other words, it’s Simply Smarter™.
The IP SAN is more flexible than a traditional Fibre Channel SAN. It

communicates with host application servers using the iSCSI protocol over
Ethernet, rather than Fibre Channel, allowing for less costly Ethernet switches
and tighter integration with an existing network.
The IP SAN connects the components by switched Ethernet connection, allowing

drive capacity and Storage Controller Modules to be managed independently.
The IP SAN is the perfect solution for decentralized, growing, and dynamic
storage needs. It includes a scalable number of IP-addressable Storage Controller
Modules and a scalable number of IP-addressable disks, which form a logical
unit called a Realm.
A Realm is an administrative domain made up of a set of designated Storage

Controller Modules, Disk Enclosures, associated software and network
interconnects. The Storage Controller Modules within a Realm are managed by
the StorControl software as a single pool of controllers, except in cases of single-
Storage Controller Module systems. The storage space provided by the Disk
Enclosures within a Realm is managed by the StorControl software as a single
virtual pool of storage space. The Realm is the entirety of the managed system:
each part of the system comes together to become the Realm. You manage the
Realm as a single entity by managing its components.
Storage traffic moves from your Initiator to the IP SAN via an Ethernet (GigE)
switch. By having a centrally-managed system, you can create easily managed
volumes for your users. Here’s how it works:
Small Computer System Interface over IP (iSCSI) uses standard IP-
addressable Ethernet connections for interaction between components.
The Initiator uses a standard network interface card (NIC) with a
software iSCSI driver OR a hardware-based HBA with driver.
SC5500 Storage Controller Modules send requests and responses through a GigE
(Ethernet) switch, handling traffic to the Initiator and the Disk Enclosure.
Acting together, the Intransa modules shown in Figure above create an

intelligently managed, single, unified system of interactive parts. The benefits of
this are:

Shared arrays of inexpensive disks

A modular, shared processor pool, simplifying shared storage; storage is
added and removed as needed
A switched GigE Ethernet infrastructure
Unified management simplifies IP SAN administration
Both graphical and command line-based management interfaces
High availability: system-wide failover is handled automatically with
minimal need for administrator intervention.
Solution Specific Configurations

Virtual volumes are configured on the Intransa Large Scale IP SAN one for
Backup for NetBackup Server named “netbackup” and other named
“source1” to “source5” on NetBackup Client as a source for data operations in
such a way that they will occupy the separate set of disks.
To setup the volumes follow the following steps:
Step 1: Create policy for creating the volumes

To create policy:
Logon to StorControl command line using the realm mgmt IP

Create a Policy using the below
StorControl> Policy create name = NTFS-Optimized virtualDiskType

= RAID10 numberOfStripeMembers = 3 stripeElementSize = 64
Step 2: Volume Creation

Backup Volume (named “netbackup”)
Volumes can be created using graphical user interface (GUI) or by using the
command line interface. Both of these methods are demonstrated one by one.
Part I: Volume Creation using CLI.
StorControl> volume create blockSize = 512b policy = NTFS-

Optimised size = 100GB name=netbackup virtualDiskplacement
= new
Response: Successful

Part II: Volume Creation using GUI

Logon to StorControl GUI using the realm management IP, “admin” user
and appropriate password.
Create a Volume using the below

Right click on All Volumes for popup menu

Use iSCSI compliant naming convention for volumes

Set RAID level through policy
Select correct unit for desired volume size
Placement - performance versus space efficiency
512-byte block size best for most applications on Microsoft Windows platform
Initialize zeros-out blocks volume is created on

Assign initiator to the Volume
Confirm new volume parameters

Volume is created and given an iSCSI Qualified Name (IQN)
Now go to the NetBackup Server (“Rasgulla”).

Install the Microsoft iSCSI Initiator software.
Double click on the “Microsoft iSCSI Initiator” icon
Available Targets will show the assigned target status as “Inactive”

Log on to the Target.
After log on you will see the target status changed to “Connected”

Now open the Computer Management window

Windows File System Tuning

Windows system create a 63-sector hidden are on disk to house the Master Boot
Record (MBR). This cause all I/O in the file system made on the subsequent
partition to be misaligned ( not on a 256KB boundary), and therefore I/O
performance will suffer. By using a program called diskpar,available on the
Windows Resource kit, you can alter the number of hidden block to a larger
number so that the disk array I/O are aligned and better optimized.
We now have initialized basic raw disks. Query the partition information using
the diskpar “i” command. (In this example, we installed the diskpar utility in a
“Windows utility” directory on the C: drive.) This indicates the disk is accessible.
C:\Documents and Settings\Administrator>diskpar -i 1

---- Drive 1 Geometry Information ----
Cylinders = 13054
TracksPerCylinder = 255
SectorsPerTrack = 63
BytesPerSector = 512
DiskSize = 107372805120 (Bytes) = 102398 (MB)
End of partition information. Total existing partitions: 0
C:\Documents and Settings\Administrator>diskpar -s 1

Set partition can only be done on a raw drive.
You can use Disk Manager to delete all existing partitions
Are you sure drive 1 is a raw device without any partition?
(Y/N) y

Cylinders = 13054
We are going to set the new disk partition.

All data on this drive will be lost. Continue (Y/N)? y
Please specify starting offset (in sectors): 64

Please specify partition length (in MB) (Max = 102398):
102398
Done setting partition.

---- New Partition information ----
StatringOffset = 32768
PartitionLength = 107372085248
HiddenSectors = 64

PartitionNumber = 1
PartitionType = 7
You now should use Disk Manager to format this partition
Verify the partition information. (Notice the starting offset is 32768 and the
number of hidden sectors is 64.)
C:\Documents and Settings\Administrator>diskpar -i 1

Cylinders = 13054
---- Drive Partition 0 Information ----

StatringOffset = 32768
PartitionLength = 107372085248
HiddenSectors = 64
PartitionNumber = 1
PartitionType = 7
End of partition information. Total existing partitions: 1
Now right click on the new volume and select the following options one by one

Mark Partition as active.

Change Drive letter and Paths and select the appropriate drive letter.
Format : Format with
o Volume Label : NetBackup Volume
o File System : NTFS
o Allocation unit size: Default
Perform a quick format
In the similar fashion create volume “Source1” to “Source5” on the IP SAN and
assign it to NetBackup Client (Jupiter)
StorControl>
StorControl> volume create blockSize = 512b policy = stripe

size = 10GB name=source1 virtualDiskplacement = new
Response: Successful

6. Installing Intransa VSS Provider

Successful Installation of Windows 2003 Operating systems with Gigabit
Ethernet NIC and Microsoft iSCSI initiator will lead to installation and
configuration of Intransa VSS Provider software on the NetBackup Client
machines (Jupiter). The Intransa VSS Provider is instrumental in talking to
Intransa IP SAN for creating snapshots of the source production volumes.
Backup will be performed from the snapshot and after the backup snapshot will
be automatically removed from the IP SAN.
Copy the Intransa VSS Provider software.

Double click on the executable.
Click Next
6. Installing Intransa VSS Provider - 34

Enter the User name and Organization
Click Next

Click Install

Click Finish. Now You have successfully installed Intransa VSS Provider
This will create the icon on the desktop.

Now double click on it for Intransa VSS Configuration.
Specify the Startup type as Automatic
Specify the Realm iSCSI target IP address, its User account and password.
Click Add realm.

Click OK
Congratulations!! You have successfully completed the Installation and

configuration of Intransa VSS Provider for the target Realm

7. Installing NetBackup Server and Client

The volumes are available for Veritas NetBackup installation and configuration.
We presume the VSS installed successfully without any error. Up until now we
have configured netbackup volume and source volumes on the Intransa IP SAN.
Also assigned them to hosts and formatted them for use. You have also
successfully installed and configured the Intransa VSS provider software on NC.
The following section explains about how to install Veritas NetBackup Software.
Now we are going to install the Veritas NetBackup Software and configure the
server and clients. Following are the simple steps to achieve this goal.
Installing the Veritas NetBackup Software.

Put the Veritas NetBackup Software media in the NS.
Click on the NetBackup Installation.
7. Installing NetBackup Server and Client - 39

Click on Install Server Software.
Click OK

Click on “I accept the terms of the license agreement.”
Select “Install to this computer only” and click “Next”.

Enter the License Key and Select the “NetBackup Master Server.”
Select “Custom” and click “Next.”

Select the “NetBackup, NetBackup Debugger Symbols” and select all

options under it.
Note: Do not select the “Veritas Snapshot Provider” option since we are going to
use the Intransa VSS Provider software for taking the snapshots.
Click Next

Select Use the “built-in system account”, In Startup select Automatic and
select “Start NetBackup Server Services After Install”.
Note: Do not modify the Default Ports except under unusual circumstances.
Click Next

Now Enter the “Master Server Name.” In this case it is “rasgulla” configured as
NetBackup Server.
Click Next
Click Install

Click Finish
This completes the installation of the Veritas NetBackup Software.

Now the NetBackup Configuration wizard will start automatically. This will help
us in configuring the NetBackup Server.
Click Next.
Click Next.

Click Next.
In Device Hosts select “rasgulla”

Click Next.
Note: Since IP SAN disks are not come in to the category of removable media
devices, Device Configuration Wizard will not find it and gives following error
Click OK.

Click Next.
Click Next.

Click Next.
Click Finish.

Click Next.
Click Next.

Click Next.
Click Next.
Please specify the full path for the Catalog backups are to be saved

Click Next.
Click Next.

Click Finish.

Now configuration wizard will start with Backup Policy Configuration.
Click Next.
Click Next.

Click OK.
This completes the Veritas NetBackup Server Software installation.
Now we have start with the Veritas NetBackup Client Software installation on
NetBackup Client machine Jupiter

NetBackup Client Installation

The following section explains about how to install Veritas NetBackup Software
on NetBackup Client machine .Now we are going to install the Veritas
NetBackup Software and configure the client. Following are the simple steps to
achieve this goal.
Installing the Veritas NetBackup Software.
Put the Veritas NetBackup Software media in the NS.
Click on “NetBackup Installation”

Click on “Install Client Software”
Click OK

Click Next
Click on the I accept the terms of the license agreement.

Click on the Install to this computer only. Click Next
Select “Custom” Click Next
Note: Do not select the “Veritas Snapshot Provider” option since we are going to
use the Intransa VSS Provider software for taking the snapshots.

Click Next

Click Next
Click Next

Click Next
Click Next
Click Install

Click Finish
This completes the Veritas NetBackup Client Installation. In the next section we
will test the backup and restore operations.

8. Testing Backup and Restore

Successful Installation of NetBackup Server and Client software will lead to
define the storage units and proper policy for taking the backup from the
snapshot.
While configuring NetBackup Server please take maximum care of defining

proper policy, storage units, Retention etc.
Defining Storage Unit
Right Click on “Storage Unit“
8. Testing Backup and Restore - 66

Select “New Storage Unit“
Fill in the relevant details and click OK

Similarly you can create multiple disk storage units to suite your backup
needs.

Defining Backup Policy

Successful Installation of NetBackup Server and Client software along with
defining storage units will lead to define the proper policy for taking the backup
from the snapshot.
Give appropriate Policy name eg. snapshot1
Click OK

Click Next
Click Next

Specify clients for this policy. Click Add and enter the name of client
Jupiter and click Next.

Click Next
Now specify files and folder to backup. Click Add and enter drive name
eg. D:

Define Backup Type , viz. Full Backup , Incremental Backup and User
Backup.

Note: Please refer to appendix for more details about the backup types
Specify the Rotation. Click Next
Specify the time window during which backups are allowed to start
Click Next

Click Finish
Now double click on the snapshot1 policy and mark the box “Perform
snapshot backups” for performing the backup operations from snapshots.
Also mark the “Allow multiple data streams” for performing multiple
backups

This is an important step in defining snapshot based backup method

Click on “Clients” tab to see the clients

Click on “Backup Selections” tab to see the volumes to be backup

Click OK. Now it will validate the policy. It will give an error is policy is
not proper.

Now click on the newly created “snapshot1” policy and click on “Manual
Backup” to start the backup immediately.
Select “Schedules” and “Clients” and click OK
Click OK

Now click OK to open the “Activity Monitor”.
Now monitor the progress of the job using the Activity Monitor on
NetBackup Server.
Following Job Details window will give Job Overview and Detailed
Status as shown below

Verifying the Snapshot operations

The snapshot operation can be verified from the Intransa StorControl window.
Following steps and screenshots demonstrate the same.
Open the StorControl window

Click on “source1” volume . In General tab section shows the camera type
icon on the volume. Indication the snapshot is initiated.
All snapshots tab shows the snapshot volume for source1 volume
connected to Jupiter in the Assigned initiator tab.

Congratulations!! You have successfully taken the backup from the snapshot
volume. Following screen shot shows the backup operation is successful.

This will verify the snapshot operation. After the Backup completes the snapshot
volume automatically gets deleted.
Similarly you can create the multiple policies to suit your backup needs.
Note: Please refer to Appendix for guidelines for defining the backup policies.
Restore Operation
The Backup data can be restored using the Backup, Archive and Restore Window
Open the “Backup, Archive and Restore” Window on NetBackup Client

or Server
Select the folders or files to be restored and click on icon to Start

Restore of Marked files.
If the data is backed using the snapshot policy then we need to restore it
to the source volume using the option Restore everything to a different
location and specifying the appropriate Destination Path

Click on “Start Restore” button.

Click on “View Status” button to monitor the progress.

Monitor the progress of the job using the “Activity Monitor” on

NetBackup Server.
Congratulations!! You have successfully taken restored volume. Following

screen shot shows the restore operation is successful.


9. Performance Benchmarking
Following performance tests were conducted on the IP SAN.
No. Test Cases Average Peak speed

speed (Mb/sec)
(Mb/sec)
A NetBackup Server & Client connected using LAN and NetBackup
Server & Client are connected to IPSAN using Intel Gigabit NIC.
1 Backup from Local Disk
Backup of NetBackup Servers 34.775 39.500
local disk using Localdisk
policy to IP SAN (NetBackup
Volume) on NetBackup Server
2 Backup from Snapshot Volume
Backup of Clients data (Source 26.441 30.500
Volume) from the snapshot on
IP SAN using Snapshot policy
to IP SAN (NetBackup Volume)
on NetBackup Server
3 Backup from Source Volume
Backup of Clients data (Source 30.153 35.554
Volume) on IP SAN using
Regular policy to IP SAN
(NetBackup Volume) on
NetBackup Server
4 Restore to IP SANDisk
Restore from NetBackup 18.567 23.045
Volume to IP SAN Source
Volume
9. Performance Benchmarking - 86
Case Study: Comparison of Backup Performance with Tape Media
No. Media Mfg. Xfer Rate Realistic Xfer 25 Gb Backup

Rate
1 4mm DDS3 02.4 Mb/s 01.2 Mb/s 5.80 hours
2 4mm DDS4 04.8 Mb/s 02.4 Mb/s 2.90 hours
3 AIT-1 06.0 Mb/s 03.0 Mb/s 2.30 hours
4 AIT-2 12.0 Mb/s 03.0 Mb/s 2.30 hours
5 DLT4000 03.0 Mb/s 01.5 Mb/s 4.60 hours
6 DLT7000 10.0 Mb/s 05.0 Mb/s 1.40 hours
7 DLT8000 12.0 Mb/s 06.0 Mb/s 1.20 hours
8 SDLT220 22.0 Mb/s 11.0 Mb/s 0.65 hours
9 SDLT320 32.0 Mb/s 16.0 Mb/s 0.45 hours
10 LTO 32.0 Mb/s 16.0 Mb/s 0.45 hours
11 Intransa IP 5500 50.0 Mb/s 38.0 Mb/s 0.15 hours
(Source: http://www.finitesystems.com/PRODUCT/tape/backup.htm)
The above chart was taken from an online web resource; it is only intended to
convey the difference between using tape as a backup medium and using the IP
5500 as a backup medium. Values represented may vary. The chart does give a
general comparison in the difference of times to perform backup operations
between tape and the IP 5500.
9. Performance Benchmarking - 87
10. Summary
Based on solution results, following are the conclusions drawn:
Conclusions
NetBackup Server with IP SAN gives Backup peak performance of 39.500
Mb/sec with average of 35.554 Mb/sec.
NetBackup Media Server with IP SAN gives Restore peak performance of
23.045 Mb/sec with average of 18.567 Mb/sec.
Disk Storage configuration is flexible and can be easily tuned to the
available disk space.
Intransa VSS Provider works well with the NetBackup Software makes it
possible to automatically start the snapshot for backup. The backup
happens from the snapshot volume and Original source volume is
available and remains online for the regular use. This helps immensely in
shorting the backup window of mission critical volumes without bringing
them offline for backup.
Restore operation can be controlled using the option Restore everything to a
different location to make sure the data is restored at the original source
volume or the required destination.
Summary
IT organizations can increase their productivity and decrease costs by
implementing a disk-to-disk backup solution. The Intransa IP 5500 provides a
robust platform for easily implementing and managing backups to disk. Veritas’s
NetBackup software was tested and performance difference was minimal for
backup with a varying difference in restore times. Together it offers the
extremely flexible and high performing fast backup solution for the enterprise.
Veritas NetBackup Software seamlessly gets integrated with the Intransa VSS
Provider. This gives immense benefit for configuring the Disk-to-Disk Backup
Solutions with Intransa IP SAN in using the automated snapshot feature.
10. Summary - 88
The volume configuration tested under Windows 2003 with ATA internal disks
for the source and an Intransa stripe of mirrors (RAID10) as the target for the
backups. The IP 5500 was configured for optimal throughput based on initial
testing by creating a stripe width of 64k per stripe member.
The IP 5500 performed much better when compared to conventional tape media.
It is important to note that only a single volume was used for the storage pool. If
multiple volumes and backup streams were used then this number could
potentially double being limited by only the software or source data read rate.
From this minimal testing, the IP 5500 decreased backup times by 66% which
drastically reduces the backup window for administrators. This has the added
benefit of allowing IT staff to increase the numbers of backups or simply ease the
management and cost of backups. This also insures that the backups are always
online and easy accessible.
For more information on specific configurations please contact your local Sales
Representative or visit http://www.intransa.com
10. Summary - 89
11. Appendix
A.VSS Concepts
This section provides theory details of the technical concepts about Microsoft
Virtual Shadow copy Service (VSS) .
A.1 What is VSS:

Microsoft’s Volume Shadow-Copy Service (VSS) provides a framework for
creating a point-in time copy of a single volume or multiple volumes. It
specifically aims to eliminate the problem of corrupt snapshots or backups due to
open files.
The Volume Shadow Copy Service (VSS) is an infrastructure that makes possible
enhanced data protection though high fidelity backups, rapid data restores, and
data transport.
VSS is a component of the operating system, and as such, is not directly accessed
by users. Instead the Volume Shadow Copy Service coordinates with user
applications, backup applications and storage hardware to enable the creation of
point-in-time shadow copies of data on single or multiple volumes without
significantly impacting performance.
Open files have historically been a backup administrator’s biggest challenge, as

they are often skipped or corrupted during the backup process. The alternative –
forcing open files closed or asking users to log out so applications can be shut
down – is no longer viable. Microsoft Windows Server 2003 VSS partially
addresses this problem by allowing users to access certain files while a backup is
taking place, without compromising the fidelity of the backup. (A drawback
to VSS is that the files must be related to a VSS-aware application in order for all
volumes to be synchronized to a single point in time,) Another feature of VSS is
the ability for users to view and recover previous versions of their files without
IT intervention.
11. Appendix - 90
By enabling read-only, point-in-time copies of an entire volume, the problem

associated with backing up files that are open or in use would seem to be taken
care of. That is Microsoft’s intent, and in some situations, it is an intent that has
the potential to be fulfilled.
A.2 VSS Components

Requestors
o Invokes VSS to a create shadow copy
Backup applications
Shadow copy management applications
Writers – Represents Apps and Windows Services
(i.e., SQL, Exchange, AD, etc.)
o Differentiates VSS from competitors
o Participate in shadow copy creation process
Providers
o Create shadow copies
System provider
Hardware snapshots
A.3 How VSS works

Microsoft Windows Server 2003 VSS works with applications that are in use to
determine when a volume shadow-copy (snapshot) can be made. A snapshot is a
copy of some data set, for example, a disk volume, at a point in time. VSS then
communicates with the operating system and applications and freezes
computing tasks to perform “copy on write” tasks on a sector-by sector basis.
Without the freeze function, data could become corrupted when the snapshot is
taken because the system might be in the middle of a task. Meanwhile,
applications continue to run uninterrupted. The original volume continues to
change as the process continues, but the shadow copy of the volume remains
constant. VSS also gives permission to the backup application to access the
volume and back it up in its “frozen” state. The shadow copy volume is then
used for the actual backup. After the shadow copy volume is saved on the
backup device, the shadow copy is deleted.
The process uses free disk space on any NTFS volume (the file system used by
Windows NT for storing and retrieving files) to store a record of the differences
between the original volume and the shadow copy volume. If there isn’t enough
temporary disk space, the volume shadow copy cannot be completed and the
backup program skips open files.
11. Appendix - 91
In order to ensure absolute data integrity for all volumes, especially those that
are interrelated, the application, the backup program and the backup device
should be “VSS aware.” This means that the components participating in the
backup must be specifically programmed to work with VSS. Towards this end,
Microsoft is providing a software development kit (SDK) to enable independent
hardware and software vendors to interface their components, including
applications, data stores, snapshot engines, and backup applications, with VSS.
These components (the application, backup program, and storage
software/hardware) are called writers, requestors, and providers, respectively, as
described in more detail in the table below.
VSS Component Description
Requestor The requestor is backup software that initiates the creation

and destruction of the shadow copy.
Writer The writer is any application software that participates in the
shadow copy process and whose files and data are included in
the process.
Provider The provider is the storage technology that does the actual
shadow copy (snapshotting).
Table: VSS Components

All components of VSS --writers, requestors, and providers-- must be VSS-aware
in order for VSS to be able to orchestrate the communication among them.
The following diagram shows the three-way orchestration among shadow copy-
writers, requestors, and providers, as performed by VSS.
11. Appendix - 92
B. Overview to VSS and VDS Design

There are two approaches to managing storage networks. One is to manage
storage through some mechanism on the server; the other is to manage it through
a separate storage product that interfaces with some hardware component of the
storage network. Both VSS and VDS are designed to enable host-based and
hardware-based solutions, as shown in Figure below. In this figure, Windows
host software components are unshaded; third party vendor components are
shaded. The additional details of this diagram are explained in later sections.
Note that some of these components might not be present.
11. Appendix - 93
VSS and VDS Components in Windows Server 2003
Software Providers
Software solutions implemented on the host computer are referred to variously
as “host-based” or ”software” solutions, or as “software providers.” Software
providers can be supplied by third party vendors, or by Microsoft (and in some
cases, both are present). In the latter case, this software is referred to as the
“system software provider,” and is an “in-box” solution that ships with the
product without requiring any additional installation steps.
The advantage of host-based solutions is that they are inexpensive and easy to
deploy. Updates and bug fixes are much simpler to manage, since they are part
of the operating system service packs. However, host-based solutions do have
drawbacks: one is that in high workload settings, the host can be overloaded and
performance impaired; another is that advanced functionality might not be
provided.
Hardware Providers
Hardware vendors have also developed their own storage management
solutions. These solutions, known as “hardware providers,” can generally
provide higher performance capabilities than solutions built into the system
software, since the workload is offloaded to the external disk subsystem.
Additionally, these solutions can provide maximum configuration flexibility, as
well as advanced capabilities such as remote management.
11. Appendix - 94
C. Volume Shadow Copy Service
The Volume Shadow Copy service provides an infrastructure for creating high
fidelity, point-in-time images known as shadow copies. These copies of a single
volume or multiple volumes can be made without affecting production server
performance. VSS is able to produce high fidelity shadow copies through its
coordination with business applications, backup applications, and storage
hardware.
Shadow copies can be used for a number of purposes:
Rapid backup and restores.
Individual file restores (using the Shadow Copy for Shared Folders
feature).
Transport (using a hardware provider) for tape backups, testing and data
mining.
VSS Components Involved in the Shadow Copy Creation

Process
Successful shadow copy creation requires the completion of a number of steps,
the details of which are quite complex. This section outlines the components of
the Volume Shadow Copy service—requestors, writers and providers—
necessary for creating high fidelity shadow copies.
11. Appendix - 95
1) Requestors: Initiating Shadow Copy Creation

The Volume Shadow Copy service is invoked by the requestor, software that
initiates basic VSS functions such as creating, importing, breaking or deleting a
shadow copy. Depending on the problem to be solved, different types of
requestor applications are invoked: backup applications for VSS-enabled
backups, management applications to manage shadow copy creation and usage,
or utility applications designed to help solve other IT problems. (One such
example is vrfydsk, which helps to verify whether chkdsk should be run).
2) Writers: Preventing Data Inconsistencies

If, during the shadow copy creation process, an application writes to disk, data
inconsistencies can occur, compromising the integrity of the point-in-time data
image. The Volume Shadow Copy service has been designed to eliminate these
drawbacks. Key to this design is the writer, an application-specific software
module that acts to ensure that an application’s data is consistent when a shadow
copy is created.
Under this design, the responsibility for data consistency has been shifted from
the requestor application to the production application itself. The advantage of
this approach is that application developers—those most knowledgeable about
their applications—can ensure, through development of their own writers,
maximum effectiveness of the shadow copy creation process.
3) Providers: Creating Shadow Copies

In the context of shadow copy creation, a provider is the interface to the point-in-
time imaging capabilities, either on either the storage array (hardware-based) or
in the operating system (software based). Windows Server 2003 includes a
system software provider with shadow copy functionality in-box; alternatively,
third party hardware and software vendors can develop their own hardware or
software providers to provide point-in-time imaging functionality. Windows
Server 2003 supports multiple hardware and software providers which can be
used in combination to solve many different IT operational scenarios.
VSS selects the provider to use during shadow copy creation using the following
default hierarchy: hardware provider software provider system software
provider. If a specific IT operational problem requires it, the requestor can
override this hierarchy.
11. Appendix - 96
Coordination of Shadow Copy Creation: the Complete

Sequence
The various roles of the requestor, writer and provider are put into context in
this section, which lists the steps that need to be taken to create a shadow copy.
Overall coordination of the requestor, writer and provider is under VSS control.
1) The requestor notifies the VSS coordinator service to prepare for shadow
copy creation.
2) The VSS coordinator notifies the application-specific writer to prepare its
data for making a shadow copy.
3) The writer prepares the data in whatever way is appropriate for that
application, such as completing all open transactions, rolling transaction logs,
and flushing caches.
4) When the data is prepared for shadow copy creation, the writer notifies the
VSS coordinator.
5) The VSS coordinator relays the message to the requestor, and the requestor
initiates the “commit” snapshot phase.
6) The VSS coordinator temporarily halts (quiesces) application I/O write
requests (I/O read requests are still possible) for the several seconds required
to create the shadow copy of the volume or volumes.
7) The provider creates the shadow copy (a maximum of 10 seconds).
8) After the shadow copy is created, the VSS coordinator releases the writer
from its temporary quiescent phase.
9) All queued write I/Os are completed.
10) VSS queries the writers to confirm that write I/Os were successfully held
during shadow copy creation.
11) If the writes were not successfully held (meaning the shadow copy data is
potentially inconsistent), the shadow copy is deleted and the requestor is
notified.
12) The requestor can retry the process (loop back to 1) or notify the
administrator to retry at a later time.
As a direct result of steps 6 and 10-11, the VSS coordinator service ensures that
the shadow copy is a highly consistent copy of the original data. Moreover,
because all the preparation for making the shadow copy is completed prior to
halting the application write I/Os, production performance is not impacted.
VSS for Fast Backups and Restores

The most common implementation of a requestor is a backup application.
Backup applications can benefit from the application-aware design of VSS for a
number of reasons. Some backup applications are designed to check the
application data—including metadata and log files—for inconsistencies. Other
backup applications do not implement consistency checks, and instead require
the administrator to ensure that applications are closed before the backup
11. Appendix - 97
process. However, even the backup applications that do complete consistency

checks have drawbacks:
Excessive Time. Depending on the amount of data involved, the
consistency check, done at the completion of a backup, can take hours. If,
by the end of the check, inconsistencies are found, the backup must be
redone. For many businesses, there simply is not the time.
Complexity. Because each application—SQL, Exchange, etc—is designed
differently, the location of the data, metadata and log files varies widely.
Backup application developers have traditionally borne the responsibility
of developing a reliable interface between the backup product and the
data to correctly find and use this information. These companies have
also been responsible for ensuring that application patches are captured
and integrated into the backup application. These fixes are not always
captured.
Using the VSS infrastructure, backup application developers can take advantage
of requestor and writer functionality. Together, these functionalities reduce the
time and complexity involved in preparing data for backup and they ensure high
fidelity point-in-time shadow copies that can be successfully and quickly
restored.
VSS for Individual File Restores

The system administrator can make shadow copies available to end users
through a feature called “Shadow Copies for Shared Folders.” The administrator
uses the Properties menu (see Figure 4) to turn on the Shadow Copies feature,
selects the volume(s) to be copied, and determine the frequency with which
shadow copies are made.
11. Appendix - 98
D. Tape Backups: Limitations

Until recently, backups to tape were the dominant technology for protecting
data. While tape backups continue to be essential for long-term archiving
solutions, they have a number of limitations that the storage manager should
consider when planning and implementing effective data protection solutions:
1. Open files cannot be backed up. Open files must either be skipped
during the tape backup process—meaning the backup data is
incomplete—or the application must be closed (“cold” backups), thereby
disrupting user productivity.
2. Tape backups are time intensive. Because of the problems associated
with backing up open files, backups are made during evening and
weekend hours. However, as businesses amass greater quantities of
mission critical data, increasingly long backup windows are required. For
many businesses, this time simply does not exist—information must be
available 24x7.
3. Tape backups are often unsuccessful. Estimates of backup failures put
the rate anywhere between 20 and 50% depending on the environment.
Worse, it is all too common that backup failures are not discovered until
the recovery process—after production data has been corrupted or
deleted. Poor backup success rates result from a number of factors,
including poor media, open files, and incomplete backups (the backup
window was too short) any of which the administrator may be unaware.
11. Appendix - 99
4. Tape restores are time intensive. Tapes must be found and retrieved
from offsite storage. While restoring a full backup is relatively
straightforward, differential and incremental backups are far more
challenging. For those backup methods to work, all the correct tapes must
be located and restored in the correct order.
5. Single file restores are costly. If an individual file is corrupted or
accidentally deleted, the user can either rebuild the file or request that an
earlier version be restored from tape. However, finding the most recent
version of a file is a non-trivial process. Because tape media is read
sequentially, the tape drive must first stream through all data physically
ahead of the sought-after file.
These limitations to tape backups can be circumvented using a relatively new
technology, variously referred to as “frozen imaging,” “point-in-time imaging,”
or “snapshotting.” Snapshots are fully useable disk-based copies that reflect the
state of the original data at an earlier point-in-time.
The Volume Shadow Copy service, new to Windows Server 2003 and Windows
Storage Server 2003, is the infrastructure that provides built-in snapshotting
(shadow copy) capabilities. Shadow copies created by Volume Shadow Copy
service augment the storage administrator’s tape backup archival solutions,
providing high fidelity point-in-time copies that can be created and restored
easily and effectively, thereby helping to simplify several aspects of storage and
data management, as explained in the following section.
High Fidelity Copies for Backup and Rapid Restores

1) Backups
Shadow copy creation through the Volume Shadow Copy service is designed to
be both an extremely fast means of creating a point-in-time image for backups—
on the order of seconds to a minute, depending on the amount of data—and a
process that produces high fidelity images without disrupting production
servers. Because shadow copy creation is non-intrusive, point-in-time copies can
be made more frequently than tape backups, thereby reducing the amount of
time a businesses data is at risk should an outage occur.
2) Restores
Scheduling non-disruptive tape backups has always been the system
administrator’s concern. Successful and rapid data restores, on the other hand, is
a concern that the entire organization shares. Unfortunately, it is also the part of
the process that most often fails. Recognizing this, organizations are taking steps
to ensure that the more complex operation of data restoration from tape, which
often must be done manually, is tested and proven effective before a problem
arises.
11. Appendix - 100

Using shadow copies to restore data from disk provides a simple and extremely
fast means of recovering data. The VSS infrastructure can be used to restore both
large amounts of data, and individual files or folders:
Large Data Restores. Shadow copies can protect single or multiple volumes from
the risks of data loss. By default, shadow copies are read-only; in the event of
data loss, they can be changed to read/write status and restored for use. Unlike
tape restores, the shadow copy restore process is very rapid—minutes rather
than hours (or more).
Individual File or Folder Restores. The Shadow Copies for Shared Folders
feature lets administrators determine which volumes to copy and to set the
frequency of shadow copy creation. The Previous Versions feature enables end
users to quickly recover point-in-time copies of individual files or folders that
have been accidentally deleted or overwritten. File recreation or lengthy single
file restores from tape are unnecessary. Instead, end users can restore earlier
versions of their own data by using a point and click process that takes only a
few seconds.
High Fidelity Backups

The biggest cause of un-restorable backups is the lack of coordination between
the backup application and end-user applications, such as databases. If
applications are open and being written to during the backup, the saved data is
inconsistent. VSS, with its shadow copy technology, solves this problem by
coordinating with backup service and the application to ensure that no writes to
disk are made during the backup process. Should there be a problem during the
shadow copy creation, VSS notifies the requestor that the shadow copy is
inconsistent, and the backup can be rescheduled. The net result of these two
factors is high data integrity.
Transport
VSS enables several advanced storage management solutions based on the ability
to transport data between servers accessing the same storage array. By design,
storage allocated to one server is not accessible by another server; if it were, the
stored data could be simultaneously written to by both servers, which would be
disastrous. In order to share data between systems, the source data must be
copied and the shadow copy made available to the second system. These shadow
copies can then be used to:
Offload backups to another server.
Make a shadow copy of production database for testing purposes.
Keep a shadow copy of data nearby for quick restore.
The net result is enhanced data protection, and better data management.
11. Appendix - 101

E. Essentials of Veritas NetBackup Administration

The NetBackup Administration Console provides a graphical-user interface
through which the administrator can manage NetBackup. The console is
available in two forms: A Windows-based interface to run on a Windows system.
A Java-based interface able to run on a Windows system. This interface is called
the NetBackup-Java Windows Display Console and is not automatically installed
on the system. Installation is available on the main NetBackup for Windows
installation screen.
Using the NetBackup Administration Console

The NetBackup Administration Console provides a Windows-based interface
through which the administrator can manage NetBackup.
11. Appendix - 102

User Backups, Archives, and Restores

To perform backups, restores and archives for this system and other clients, open
the client interface by clicking on the Backup, Archive, and Restore button in the
toolbar.
Users can back up, archive, and restore files, directories, and raw partitions that
reside on their own client computer. A user can restore files at any time but can
back up and archive only during the time periods that the administrator defines
with the schedules. Users can also view the progress and final status of the
operations they perform.
Activity Monitor
Use the Activity Monitor to monitor and control NetBackup jobs, services,
processes, and drives.
E.1 NetBackup Management

Reports
Use Reports to compile information for verifying, managing, and
troubleshooting NetBackup operations.
Policies
Use Policies to create and specify the backup policies which define the rules for
backing up a specific group of one or more clients. For example, the backup
policy specifies when automatic backups will occur for the clients and when
users can perform their own backups. The administrator can define any number
of backup policies, each of which can apply to one or more clients. A NetBackup
client must be covered by at least one backup policy and can be covered by more
than one.
The properties of a backup policy include the following:

General attributes that define the:
Priority of backups for this policy relative to backups for other policies.
Storage unit to use for backups of clients covered by this policy.
Volume pool to use for backups performed according to this policy. A
volume pool is a set of volumes that the administrator can assign to
specific backup policies or schedules. For example, it is possible to have
one volume pool for weekly backups and another for quarterly backups.
List of client computers covered by the policy.
11. Appendix - 103

List of files to include in automatic backups of the clients. The backup

selection list does not affect user backups because the user selects the
files.
Schedules that control when backups and archives can occur for the
clients.
As mentioned above, each backup policy has its own set of schedules.
These schedules control when automatic backups start and also when
users can start a backup or archive.
Each schedule is unique with attributes that include:
Type of schedule. Specify schedules for automatic full or incremental
backups or user backups or archives. There are also schedule types that
apply only when separately-priced options are installed (for example, a
backup schedule for Microsoft Exchange or Oracle databases).
Backup window. For automatic full or incremental backup schedules, this
is the time period when NetBackup can start automatic backups of clients
covered by this policy.
For user schedules, this is the time period when users can start a backup
or archive of their own client.
Frequency. How often automatic and calendar-based backups should
occur and which dates should be excluded from the schedule (dates when
backups should not occur).
Retention. How long NetBackup keeps the data that is backed up by this
schedule.
Storage unit. The storage unit for the data that is backed up by this
schedule. This setting, if used, overrides the storage unit specified at the
backup policy level.
Volume pool. The volume pool to use when saving data backed up by
this schedule. This setting, if used, overrides the volume pool specified at
the backup policy level.
The administrator can also manually start a backup schedule for an automatic
full or incremental backup. Manual backups are useful if, for example, a client
system is down and misses its scheduled backup.
Configuring Backup Policies:

The easiest way to set up a backup policy is to use the Backup Policy
Configuration Wizard. This wizard guides you through the setup process,
simplifying the process by automatically choosing default values that are good
for most configurations.
11. Appendix - 104

Policy Planning Guidelines for Backups

Policies allow you to meet the needs of a wide variety of clients in a single
NetBackup configuration. However, taking full advantage of policies for use in
backups requires careful planning before starting your configuration. The
following procedure provides planning guidelines.
1. Divide clients into groups according to the types of work they perform.
2. Gather information about each client. Include information relevant to the
backups such as the names, size, and number of files.
3. Create backup policies to accommodate special storage requirements.
4. Create additional backup policies if one set of schedules does not
accommodate all clients and files.
Best times for backups to occur.
How frequently the files change.
How long backups have to be retained.
5. Create separate policies for clients that require different general attribute
settings than other clients.
6. Create separate policies as necessary to maximize the benefits of
multiplexing.
7. Evaluate total backup times for each schedule and further subdivide your
policies to reduce backup times to an acceptable level.
When to Use Multiple Data Streams
Reduce Backup Time

Multiple data streams can reduce the backup time for large backups. This is
achieved by splitting the backup into multiple streams and then using
multiplexing, multiple drives, or a combination of the two for processing the
streams concurrently.
In addition, configuring the backup so each physical device on the client is

backed up by a separate data stream that runs concurrently with streams from
other devices can significantly reduce backup times.
Note For best performance, use only one data stream to back up each physical
device on the client. Multiple concurrent streams from a single physical device
can adversely affect backup times because the heads must move back and forth
between tracks containing files for the respective streams.
11. Appendix - 105

Setting the Number of Streams That Can Run Concurrently

The number of streams that can run concurrently for a policy or client is
determined by the following:
Storage unit and schedule multiplexing limit
Number of drives that are available
Maximum concurrent jobs settings for the policy and client
Each storage unit and each schedule has a maximum multiplex setting. The
lower of the two settings is the limit for a specific schedule and storage unit. The
maximum number of streams that can be multiplexed is limited to the sum of the
multiplexing limits for all drives available in the storage unit and schedule
combinations.
For example, assume there are two storage units with one drive in each. MPX on
storage unit 1 is set to 3 and MPX on storage unit 2 is set to 5. If MPX is set to 5 or
greater in the schedules, then 8 streams can run concurrently.
The maximum jobs settings also limit the maximum number of streams:
Maximum Jobs Per Client (Host Properties > Master Server > Global
NetBackup Attributes)
Limit jobs per policy (policy attribute)
Maximum Data Streams (Set the number in Host Properties > Master
Server > Client Attributes) The maximum job settings are interdependent
as follows:
o If Maximum Data Streams is not set, the lowest value of
Maximum Jobs Per Client and Limit Jobs Per Policy is the limiting
factor.
11. Appendix - 106

o If Maximum Data Streams is set, then NetBackup ignores

Maximum Jobs Per Client and uses the lowest value of Maximum
Data Streams and Limit Jobs Per Policy as the limiting factor.
Storage Units
Use Storage Units to display storage unit information and provide commands for
managing NetBackup storage units.
A storage unit is a group of one or more storage devices of a specific type and
density that attach to a NetBackup server. The media can be removable (such as
tape) or a directory on a hard disk. Removable media can be in a robot or a
standalone drive.
The devices in a removable-media storage unit must attach to a NetBackup

master or media server and be under control of Media Manager. The
administrator first sets up Media Manager to use the drives, robots, and media
and then defines the storage units. During a backup, NetBackup sends data to
the storage unit specified by the backup policy. Media Manager then picks an
available device within the storage unit.
When the storage unit is a directory on a hard disk, the administrator specifies
the directory during configuration and NetBackup sends the data to that
directory during backups. Media Manager is not involved.
Storage units simplify administration because once they are defined, the
NetBackup configuration points to a storage unit rather than to the individual
devices it contains. For example, if a storage unit contains two drives and one is
busy, NetBackup can use the other drive without administrator intervention.
Disk Storage Unit Considerations

A disk type storage unit consists of a directory on a hard disk that stores the
backup or archive data. NetBackup permits an unlimited number of disk storage
units. A disk type storage unit is useful for testing and is useful during busy
periods because it allows quick backups. However, you must be careful that it
does not fill up your disk.
Before using a disk storage unit, configure the disk as explained in your
operating system documentation. To calculate the approximate disk space that
NetBackup requires as it creates backups, use the following formula:
(Largest backup size x (number of backups + 1)) + Space for the restores that
are concurrent with backups
11. Appendix - 107

Catalog
Use Catalog to create and configure a special type of backup NetBackup requires
for its own internal databases—a catalog backup. These databases, called
catalogs, are on the NetBackup serverʹs disk and have setup information as well
as critical information on client backups. The catalog backups are set up and
tracked separately from other backups to ensure recovery in case of a server
crash.
Catalog is also used to search for a backup image in order to verify the contents
of media with what is recorded in the NetBackup catalog, to duplicate a backup
image, to promote a backup image from a copy to the primary backup copy, to
expire backup images, or to import expired backup images or images from
another NetBackup server.
Host Properties
Use Host Properties to customize NetBackup configuration options. In most
instances, no setting changes are necessary. However, Host Properties settings
allow the administrator to customize NetBackup to meet specific site preferences
and requirements for master server, media servers, and clients.
Media and Device Management

The software that manages the removable media and storage devices for
NetBackup is called Media Manager. This software is part of NetBackup and is
installed on every NetBackup server. The administrator can configure and
manage media through Media and Device Management in the NetBackup
Administration Console.
Access Management
Customers can protect their NetBackup configuration by using Access
Management to define who may access NetBackup and what functions a user in
a user group can perform.
11. Appendix - 108

E.2 NetBackup Configuration Wizards

The easiest way to configure NetBackup is to use the configuration wizards. The
wizard selection is visible in the Details pane on the right varies depending on
what NetBackup node is selected in the left portion of the screen.
Getting Started
Use the Getting Started Wizard if you are configuring NetBackup for the first
time. It leads you through the necessary steps and other wizards to get you up
and running with a working NetBackup configuration.
Configure Storage Devices

Use the Device Configuration Wizard to guide you through the entire process of
configuring a device and storage unit.
Configure Volumes
Use the Volume Configuration Wizard to guide you through the entire process of
configuring removable media.
Configure the Catalog Backup

Use the NetBackup Catalog Backup Wizard to set up your catalog backups,
which are essential to recovering your data in case of a server failure or crash.
Create a Backup Policy

Use the Backup Policy and Configuration Wizard to add a backup policy to your
configuration.
Import Images Wizard

Use the Import Images Wizard to import NetBackup images in a two-part
process.
Configure a Shared Drive

Use the Device Configuration Wizard to configure NetBackup to use shared
drives or to reconfigure an existing shared drive.
11. Appendix - 109

E.3 Disk Staging Storage Unit Considerations

Disk staging provides a method for administrators to create images on disk
initially, then later copy the images to another media type (as determined by the
disk staging schedule).
The later media type would typically be tape, but could be disk or disk staging.
This two-stage process allows the NetBackup administrator to leverage the
advantages of disk-based backups in the near term, while preserving the
advantages of tape-based backups for long term.
Disk staging may be appropriate for your NetBackup environment to meet the
following objectives:
To allow backups when tape drives are scarce.
To allow for faster restores from disk.
To facilitate streaming to tape without image multiplexing.
Disk Staging is conducted in two separate operations:

Stage I: A backup creates an image on the disk staging storage unit.
Stage II: A relocation schedule determines when the image from the disk
staging storage unit should be relocated to the destination storage unit.
The image continues to exist on both the disk staging storage unit and the
destination storage unit. File restores are done from the disk staging storage unit
copy, while the destination storage unit copy can be considered the long term
copy.
11. Appendix - 110

The image copy continues to exist on the disk staging storage unit until either the
copy expires based on the copy’s retention period, or until another Stage I
process needs space on the disk staging storage unit.
When a Stage I process detects a full disk staging storage unit, it pauses the
backup, finds the oldest image that has been successfully copied to the
destination storage unit, and expires this image copy.
Maximum Concurrent Jobs

For hard disk storage units, Maximum Concurrent Jobs specifies the maximum
number of backups that NetBackup can concurrently send to this disk. For
example, if there are three backup jobs for this storage unit and Maximum
Concurrent Jobs is set to two, the first two jobs start and the third one waits.
Maximum Concurrent Jobs corresponds to the Number of Drives setting for a

Media Manager storage unit. The jobs are not multiplexed.
The number to enter depends on the available disk space and the server’s ability
to comfortably run multiple backup processes.
Maximum Fragment Size

The Maximum Fragment Size setting specifies (in megabytes) the largest
fragment size that NetBackup can create when storing backups.
For Media Manager storage units:

The default maximum fragment size for a Media Manager storage unit is 1
terabyte. To specify a maximum fragment size other than the default, place a
check in the Maximum Fragment Size check box, then enter a value of 50
megabytes to 1048576 megabytes (1terabyte).
For hard disk storage units:

The default maximum fragment size for a disk storage unit is 2000 megabytes. To
specify a maximum fragment size other than the default, enter a value that
ranges from 20 megabytes to 2000 megabytes.
Fragmenting disk backups is normally used to ensure that the backup does not
exceed the maximum size allowed by the file system. It is intended primarily for
storing large backup images on a disk type storage unit. Another benefit of
fragmenting backups on disk is increased performance when restoring from
images that were migrated by Storage Migrator. For example, if a 500 megabyte
backup is stored in 100 megabyte fragments, you can restore a file quicker
because Storage Migrator has to retrieve only the specific fragment with the file
rather than the entire 500 megabytes.
11. Appendix - 111

Maximum Multiplexing per Drive

Specifies the maximum number of backups that NetBackup can multiplex onto
any single drive in the storage unit. To enable multiplexing, place a check in the
Maximum Multiplexing per Drive check box, then specify any value from 2
through 32.
E.4 Types of Backup

The Type of Backup specifies the type of backup that a schedule will control.
Click in the field and select from the drop-down list. The list displays only the
backup types that apply to the policy you are configuring.
If the schedule is a relocation schedule, created as part of configuring a disk
staging storage unit, there is no backup type selection to be made.
The following is a list of possible selections:
Full Backup
Backs up all the files that are specified in the backup selections list for the policy,
regardless of when they were last modified or backed up. These backups occur
automatically according to the criteria in the schedule. If you use incremental
backups, you must also schedule full backups to perform a complete restore. If
you’re performing a raw partition backup, you must select Full Backup.
Cumulative Incremental Backup

Backs up all files that are specified in the backup selections list and have changed
since the last successful full backup. All files are backed up if no prior backup
has been done. These backups occur automatically according to the criteria in the
schedule. A complete restore in this instance requires the last full backup and the
last cumulative incremental.
NetBackup recommends that you do not combine differential incremental

backups and cumulative incremental backups within the same Windows policy
when the incremental backups are based on archive bit (default).
Differential Incremental Backup

Backs up all files that are specified in the backup selections list for the policy and
have changed since the last successful incremental or full backup. All files are
backed up if no prior backup has been done. These backups occur automatically
according to the criteria in the schedule. A complete restore in this instance
requires the last full backup, the last cumulative incremental and all differential
incrementals that have occurred since the last full backup.
11. Appendix - 112

User Backup
Initiated by the user through the client interface (Backup, Archive, and Restore
interface) and backs up all files that the user specifies. Users can start backups
only during the times that you specify in the schedule Start Window tab.
If the schedule is to be used for a catalog archive, User Backup must be selected
for the backup type.
User Archive
Initiated by the user through the interface on the client and archives all files that
the user specifies. An archive is a special type of backup that first backs up the
file and then deletes it from the local disk if the backup is successful. This frees
local disk space while still keeping a copy for future use (until the retention
period expires). Users can start archives only during the times that you specify in
the schedule Start Window tab.
Synthetic Backups
A synthetic full backup is a backup assembled from a previous, traditional (non-
synthesized) full backup, and subsequent differential backups and/or a
cumulative incremental backup. A client can then use the synthesized backup to
restore files and directories in the same way that a client restores from a
traditional backup.
Advantages of Using Synthetic Backups

There are a number of reasons why implementing synthetic backups may be
useful in your NetBackup configuration.
Processing Takes Place on Master and Media Server(s) Instead of Client
Reduce Network Traffic
Use Drives More Effectively
11. Appendix - 113

E.5 Configuring an Advanced Client Policy

This procedure explains how to configure Advanced Client features in a policy.
1. Start the NetBackup Administration Console as follows: On Windows, click
Start > Programs > VERITAS NetBackup > NetBackup
2. Administration Console.
3. Click on Policies.
4. In the All Policies pane, double click on the name of the policy (or right-click
to create a new one).
5. Select the policy type:
If client data is in a database, select the database type (Oracle,
MS-Exchange-Server, MS-SQL-Server).
To use FlashBackup, select FlashBackup for UNIX clients or
FlashBackup-Windows for Windows 2000/2003 clients.
For all other cases, select Standard for UNIX clients and MS-Windows-NT
for Windows clients.
6. Make sure Perform snapshot backups is selected.
7. To create a backup that enables Instant Recovery, select the Retain snapshots
for instant recovery attribute. This attribute is required for the following
types of restore: block-level restore, file promotion, and image rollback.
8. To reduce the processing load on the client, select Perform offhost backup.
9. To save these settings, click Apply.
11. Appendix - 114

11. Appendix - 115

Reference and Acknowledgements
References
Intransa IP 5500 Storage Realm
http://www.intransa.com
IP SAN Administrator’s Guide

http://engr.intransa.com/bedrock/docs/Bedrock_Admin_Guide/Hokie/Hokie_SC
5500AdminGuide.pdf
IP SAN Command Line Interface Guide

http://engr.intransa.com/bedrock/docs/Cli/Hokie/SC5500CLI.pdf
IP SAN StorControl Management Tool Guide

http://engr.intransa.com/bedrock/docs/Bedrock_GUI_Guide/Hokie/SC5500_Stor
Control.pdf
Veritas Inc.
www.veritas.com
Microsoft Corporation
www.microsoft.com
Acknowledgements
We highly appreciate the efforts taken by the Intransa team in formulating, testing
and compiling this Solution Document.
Reference and Acknowledgements - 116

Intransa, Inc.
2870 Zanker Road Suite 200
San Jose, CA 95134-2114
Phone: 408.678.8600
Fax: 408.678.8800
Web: http://www.intransa.com

D2 DUsing Veritas Netbackup

Caricato da

Informazioni sul documento

Descrizione originale:

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

D2 DUsing Veritas Netbackup

Caricato da

Copyright:

Formati disponibili

Online Disk-to-Disk Backup with

Veritas NetBackup and VSS

Microsoft® is a registered trademark of Microsoft Corporation. Windows™ is a trademark of Microsoft

3. Veritas NetBackup .................................................................................................... 12

4. NetBackup Qualification Setup ............................................................................. 16

5. Configuring IP SAN Volumes................................................................................ 22

6. Installing Intransa VSS Provider ........................................................................... 34

7. Installing NetBackup Server and Client .............................................................. 39

8. Testing Backup and Restore ................................................................................... 66

11. Appendix .................................................................................................................. 90

Reference and Acknowledgements ......................................................................... 116

Backup-to-disk is emerging as powerful solution for faster recovery of mission-

Most of the applications today require frequent retrievals of recently captured

Advanced Technology Attached (ATA) disk technology provides greater benefits

Today, Intransa recommends adopting a backup-to-disk solution when

Intransa Disk-to-Disk (D2D) Backup Strategy

Intransa IntraStor™ is a 3-tiered distributed architecture that fully leverages the

This solution of Disk-to-Disk backup is using Veritas NetBackup software and

This paper gives an overview of how to deploy Disk-to-Disk Backup Solution on

The paper is lucidly divided into the following logical sections:

Section 3 provides the overview of Veritas NetBackup Software, its architecture,

Section 4 provides the overview of Veritas NetBackup Qualification

Section 6 provides step-by-step instructions for installing and configuring the

Section 7 provides step-by-step instructions for Installation and configuring

Section 11 is the appendix containing the additional information about the

This document contains information relating to configuring Veritas NetBackup

Intransa Solution Engineering teams performed tests using specific system

Disk Backup Overview

Random access characteristics of disk versus sequential access for tape

The following sections summarize these benefits.

Media Reliability and Data Availability

Intransa’s new IP SAN Storage System is a second-generation iSCSI-enabled,

Intransa Approach to Solutions

Intransa’s solution leverage on the simplicity, scalability and flexibility of the

The following approach is helping Intransa to rapidly build a growing and

up the performance on demand. Intransa will continue to develop value add

For more details about Intransa please visit http://www.intransa.com

The VERITAS NetBackup mission is to provide heterogeneous data protection

The NetBackup Enterprise Server product provides powerful functionality,

The NetBackup Server product is a high-performance solution offering ease of

For more details about Veritas please visit http://www.veritas.com

Other Useful Resources

A NetBackup Master Server is the manager of the storage domain. An

NetBackup provides high-performance backups and restores for a variety of

Administrators can set up schedules for automatic, unattended backups for

NetBackup includes both the server and client software:

About Advanced Client

Advanced Client combines the features of snapshot backup, FlashBackup, BLI

A snapshot image is required for all features of Advanced Client. A number of

A solution to this problem is to create a snapshot of the data. This means

When a backup is managed by a backup agent on a Storage network, the data to

4. NetBackup Qualification Setup

4. NetBackup Qualification Setup - 16

4. NetBackup Qualification Setup - 17

The proposed D2D solution comprises of the following architecture:

The NetBackup Server (NS) is installed on an independent server and connected

4. NetBackup Qualification Setup - 18

NetBackup Server provides scheduling and management functionality for

Configurations used in the Qualification Environment

NetBackup Server (NS)

NetBackup Client (NC)

1. LAN: Local Area Network